Running a Demo for Seven People: Who to Talk To, Who to Watch, and Who Kills It Later
A demo with seven stakeholders is a different job than a one-on-one — here are the mechanics I use, from the pre-call email to the follow-up for the exec who dropped.
A one-on-one demo is a conversation. A demo with seven people on it is a meeting, and meetings have rules that conversations do not. If you run the seven-person call the way you run the one-on-one — open the product, walk the happy path, ask for questions at the end — you will get a polite thank you, a request for pricing, and then nothing. The deal will die in a debrief you were not invited to, killed by someone whose name you never said out loud.
I have sat on both sides of this. Selling inside AWS and Dell, I ran calls where the room had a user, a manager, an architect, a procurement person, and someone from security who typed the whole time and never unmuted. And as a founder buying software four times over, I have been the guy on the call who said almost nothing and then told my team no afterward. The people who sold me were the ones who figured out, during the call, that I was the one who mattered and that I was not going to raise my hand.
So let me give you the mechanics. Not theory. The actual sequence I would run if I had a multi-stakeholder demo on the calendar for Thursday.
The pre-call email is the whole game
Everything good that happens on a committee demo was set up before the call started. Everything bad that happens is usually because you walked in blind and had to improvise a room you did not understand.
When the calendar invite comes back with names you did not expect, you send this. Same day. Do not wait.
Hi Dana — saw a few more people got added to Thursday, which is great. So I do not waste anyone's time with the wrong parts of the product, can you tell me two things?
- For each person, what is the one thing they are going to want to see? Even a few words each is enough.
- Is there anyone on the list who is going to be skeptical, and what are they skeptical about?
I would rather spend the hour on the things that matter to them than run my standard walkthrough. If it is easier to do this on a five-minute call tomorrow, I will take that instead.
Two things to notice. First, I am not asking for titles. Titles tell you almost nothing — a "Director of Operations" could be the champion or the person who has to rip out the tool they built themselves. I am asking what each person wants to see, which is a question about stake, not about org chart. Second, I am explicitly inviting the champion to name the skeptic. Champions know. They almost always know, and most of them will tell you if you ask in a way that does not sound like you are afraid.
If they answer by email, you have a script. If they take the five-minute call, you have a script plus tone, which is better. Either way you now know things like: the ops manager tried a competitor two years ago and it was a mess, the security lead has a rule about where data lives, the VP is only coming because her boss asked her to look at this category.
When you do not get a reply, do not walk in blind. Go find each person on LinkedIn and write one sentence about what you think their stake is. You will be wrong about some of them. Being wrong on purpose and then correcting on the call is dramatically better than having no theory at all, because at least you will be listening for the correction.
Ask who is missing
One more line for that email, and it is the one most reps skip: "Is there anyone who is not on the invite but will have a say in the decision?"
Someone is always missing. On big committee sales it is usually finance or a specific technical approver. In regulated industries it is almost guaranteed — the legal demo playbook exists because managing partners routinely send an associate to the first call and make the real decision in a partners' meeting you will never see. Knowing that person exists changes how you run the whole hour, because your job shifts from convincing the room to arming the room.
The first ninety seconds: say every stake out loud
Here is the single highest-leverage thing you can do on a packed call, and it takes less than two minutes.
After the hellos, before you share your screen, you name each person and what you think they care about. Out loud. To the room.
Before I show anything — I want to make sure I know why each of you is here, because I would rather cut a section than talk past someone.
Dana, you and I have talked, so I know your piece is the handoff between intake and review.
Marcus, my understanding is you run the team that would actually live in this thing every day, so I am guessing you want to know what a Tuesday looks like, not what the dashboard looks like.
Priya, I have not met you — I assumed you are here on the data side. Is that right, or is there something else you need out of this?
Tom, same question. Dana mentioned you would want to look at how this touches the existing system.
And Reena, I know your time is tight, so if there is one thing you need answered before you drop, tell me now and I will make sure we hit it in the first twenty minutes.
Read that again and count what it does. It tells everyone you did homework. It gives the quiet people a low-cost way to correct you, which is much easier than asking them to volunteer their concerns cold. It surfaces the exec's real question before she leaves. And it makes you the person running the meeting, which — on a call with seven people and no agenda — is a role nobody else is going to take.
The corrections you get here are gold. When Priya says "actually I am here from security, not analytics," you just avoided forty minutes of showing her the wrong product. When Tom says "honestly I am mostly here to understand what this replaces," you now know he is measuring you against something he already owns, which is a completely different conversation than a greenfield one.
You will feel like you are burning time. You are not. The demo you give after those ninety seconds is a different demo.
One question per stakeholder before you show anything
The rule I hold to: nobody sees the product until every person in the room has spoken at least once.
This is not a warmth exercise. It is a risk exercise. The person who does not speak in the first fifteen minutes will not speak in the last fifteen either, and their objection does not disappear — it just moves to the debrief where you cannot answer it.
So you go around. Short questions, specific to the stake you just named.
To the user: "Marcus, walk me through what happens today when one of these comes in after hours." To the technical person: "Tom, what does the current integration look like — is that something your team built?" To the risk or security person: "Priya, what is the thing that usually kills a tool like this in your review?" To the exec: "Reena, if this works, what changes for you?"
That security question is deliberately blunt and I ask it almost every time. People in risk roles spend their working lives being routed around, and asking them directly what disqualifies a vendor tends to get a straight answer. Sometimes the answer is "we cannot use anything that stores data outside our region," and you find out in minute nine instead of week six.
Budget your time for this. On a sixty-minute call I would rather spend the first twenty on the roster and the questions and then show a tight, targeted twenty-five than run a fifty-minute product tour. Nobody has ever complained that a demo was too focused on their problem.
The quiet person is the risk
Every committee has one, and they are usually in risk, security, compliance, or IT. They will be pleasant. They will not push back. They will type notes. And then in the debrief they will say one sentence — "I have concerns about how they handle X" — and that sentence carries more weight than everything your champion says, because it is a no, and in a room full of maybes a no wins.
You cannot out-charm this person. You have to give them something to do.
The move is to hand them a deliverable during the call. "Priya, I do not want to guess at your review process. What if I send you our security documentation and the data flow diagram this afternoon, and we book fifteen minutes just you and me next week to go through it? I would rather find the blockers now than in procurement." You have just converted a silent veto into a scheduled meeting where you get to answer objections directly instead of hearing about them secondhand.
Carrier buyers do this to you constantly — the insurance demo playbook is built around the fact that claims and compliance people watch a demo specifically hunting for the place it breaks, and pretending otherwise just makes you look naive to the one person you needed to respect. Same in utilities, where the asset risk demo for a distribution network usually has an engineer on it who is quietly checking whether your model does anything their spreadsheet does not.
When the quiet person asks a hard question, do not soften it back to them. Answer it, and if you cannot, say so plainly and commit to a date. "I do not know. I will find out and have an answer to you by Friday." Technical skeptics have heard every dodge. A straight I-do-not-know buys you more credibility than a smooth non-answer, every single time.
Keeping the senior person engaged without ignoring the users
The trap is that these two audiences want opposite things. The users want to see clicks. They want to know what Tuesday morning feels like, whether the thing is fast, whether it makes their job worse before it makes it better. The executive does not care about clicks and will start checking email the moment you start clicking.
What I do is narrate for the exec while demoing for the users.
You show the actual workflow — real screens, real speed, Marcus's Tuesday morning. And every couple of minutes you surface, for one sentence, to say what it means. "So that is the flow. Reena, the thing to notice there is that nobody had to leave the system to get an approval — that is the piece Dana said adds two days right now." Then straight back down into the product.
That one sentence is doing translation work. It also keeps the exec's attention because you are addressing her by name, which is very hard to ignore even when you are half-reading email.
The other half of this is that you must let the users talk while the exec is on. Executives read the room. If Marcus says "oh, that would save my team the Friday reconciliation," that is worth ten times more to Reena than you saying the same thing. Your job is to create the conditions where the user says it in front of the boss. So when a user reacts, stop and pull the thread: "Say more about the Friday thing."
When the executive drops at minute fifteen
She will. Plan for it.
When it happens, do not pretend it did not. Take ten seconds and say it out loud: "Okay, Reena is off. Dana, I will make sure she gets a short version of what she missed. Let's use the rest of the time to go deeper on the operational side, since that is where the real questions are."
This does two things. It signals to the remaining people that you know the exec matters and you are handling it, and it visibly changes the meeting into a working session, which is what the users actually wanted anyway. The energy usually goes up after the senior person leaves. Let it.
Then — and this is the part people skip — before the call ends, ask the room what to send her. "If you were briefing Reena in the hallway tomorrow, what would you tell her about this?" You will get the room's own summary in the room's own language, and now you know exactly what your follow-up needs to reinforce and where the story is weak.
The follow-up is three emails, not one
One recap email to seven people is a document nobody reads. Here is what I send instead, same day.
To the champion, the long one. What you heard from each person, what you think the open items are, who you think the risk is, and a direct ask: "My read is Priya is the one we need to satisfy before this can move. Does that match yours?" Champions will correct you, and their correction is a map of the internal politics.
To the skeptic, a separate note with the specific thing they asked for and a calendar link. Short. No marketing language. Just the document and the fifteen minutes.
To the absent or departed executive, something built for her and nobody else. Not the recap. Four or five sentences: what her team said, in her team's words, the one number or outcome they cared about, and the decision she needs to make next. If you can, attach a three-minute recording of just the section she missed — not the full call, the section. Almost nobody does this, and it is the closest thing to a free advantage in committee sales, because it means the person with the most power gets a version of your pitch made for her rather than a forwarded meeting she will not watch.
Then send the shared recap to the group so the record exists and your champion has something to circulate.
The part you actually have to practice
All of this fails in the moment for the same reason: the first ninety seconds are hard to do cold. Naming five people and their stakes, out loud, without notes-reading stiffness, while your screen share is not up yet and nobody has warmed up — that is a performance, and performances need reps. So does saying "I do not know, I will get you an answer by Friday" without the little apologetic upswing at the end that undoes it.
What I would do next is take your last packed demo, write down the five people who were actually on it, and run the opening out loud against each of those names until it sounds like you are thinking rather than reciting. If you want a room that pushes back the way a real committee does — the quiet security lead, the exec who drops, the user who liked the old system — that is what we built DrillCall for, and the committee demo is the scenario I would drill first, because it is the one where a single unspoken objection costs you the whole quarter.
The demo is not the hard part. Seven people with seven different reasons to say no is the hard part. Name them, make them talk, and make sure the one who was not in the room gets a version made for them.